Computing system performance is generally dependent on the proper management of the execution of software processes (also referred to as jobs, or tasks). In small scale systems, managing the execution of processes is relatively straightforward and can be normally accomplished through a first-in, first-out (FIFO) method for organizing said processes. In small scale systems, the number of processes is typically small and the computing resources required are nominal. Consequently, the amount of delay in completing a process is insignificant and does not adversely affect system performance.
In large scale systems with processes requiring significant computing resources such as in a data collection system which retrieves, stores and manages large databases for data mining purposes, a queuing method such as a FIFO can substantially impact throughput performance of critical processes when all processes including non-critical processes are given equal access to computing resources.
A need therefore arises for a system for managing data collection processes.